Hi Everyone,
I am trying to determine how the mean diff is computed. Here is the
output from my analysis.
Call:
matchit(formula = treat ~ C4549921864, data = obs_data, method = "nearest",
distance = "mahalanobis")
Summary of balance for all data:
Means Treated Means Control SD Control Mean Diff eQQ Med
C4549921864 1 0.7205 0.4502 NA NA
eQQ Mean eQQ Max
C4549921864 NA NA
Summary of balance for matched data:
Means Treated Means Control SD Control Mean Diff eQQ Med
C4549921864 1 1 NaN NA NA
eQQ Mean eQQ Max
C4549921864 NA NA
Percent Balance Improvement:
Mean Diff. eQQ Med eQQ Mean eQQ Max
C4549921864 100 NA NA NA
Sample sizes:
Control Treated
All 161 1
Matched 1 1
Unmatched 160 0
Discarded 0 0
How come the mean diff is NA? I thought it should be the difference between
the treated and control means.
Thanks
George
Hello,
I am encountering two issues with MatchIt and would appreciate any advice
that people can offer. I am fairly new to this and so I'm sorry if I am
making an obvious mistake. I have not been able to find an answer to my
questions in the MatchIt documentation.
1) I am trying to implement nearest neighbor matching with a caliper
restriction. Each time that I run the code, I get slightly different
results for the balance of the matched data (the mean for the matched
control group is slightly different each time that I run the code). This
does not happen when I implement nearest neighbor matching without a
caliper restriction.
Here is a piece of my code.
`mydata` <- read.dta("mydata.dta")
m.out <- matchit(treat ~ x1+ x2 + x3+ x4 + x5, data = mydata,
method="nearest",caliper = .1)
summary(m.out, standardize=T)
2) I would also like to try optimal matching. I keep encountering the
following warning message after I try it.
"Warning message:
In fullmatch(d, min.controls = ratio, max.controls = ratio, omit.fraction =
(n0 - :
Without 'data' argument the order of the match is not guaranteed
to be the same as your original data."
Here is my code for optimal matching:
m.outopt <- matchit(treat ~ x1+ x2 + x3+ x4 + x5, data = mydata,
method="optimal")
summary(m.outopt, standardize=T)
Any advice?
Thank you!
Barbara
Dear all,
I completed a matching procedure in R with the package Matchit, using the Full matching option. Now I would like to carry out a sensitivity analysis to measure the magnitude of potential unmeasured covariates that would need to be present to alter the study's conclusions.
The package rbounds only allows one-to-one or one-to-two/three matching... With full matching, matched sets differ in size. Is there a way to conduct a sensitivity analysis manually after full matching?
Another thought is to extract subsamples of each of the matched sets , and to do this as many times as there are control units it the largest set. Then, to use the rbound package and repeat the analysis for each subsample. But this is kinda cumbersome and might not be totally addressing sensitivity analysis in essence.
Any help would be great!
Thanks.
Best,
Machteld Vandecandelaere
Machteld Vandecandelaere
University of Leuven - Centre for Educational Effectiveness and Evaluation
Dekenstraat 2, box 3773 - room VHI 2.43
B-3000 Leuven
Tel: + 0032 16 325 780
E-mail: Machteld.Vandecandelaere(a)ppw.kuleuven.be<mailto:Machteld.Vandecandelaere@ppw.kuleuven.be>
URL: http://ppw.kuleuven.be/o_en_o/COE
Hi All,
I am trying to compute the balance on my data using matchit from a
function.
computes_balance <- function(obs_data){
obs_data = as.data.frame(obs_data);
//compute formula_str here ...
matchit(formula_str, data = obs_data, method = "exact");
}
Matchit always gives me this error message :
$ operator is invalid for atomic vectors
But when I run matchit from the command line, I don't get this error.
Any idea as to what is happening?
Thanks.
George
Try 3.5-5. It works for me.
Kosuke
Department of Politics
Princeton University
http://imai.princeton.edu
On Feb 11, 2013, at 2:05 PM, Antonio Bubbico <antonio.bubbico(a)gmail.com> wrote:
> Hello,
> I have the same problem Mao had, and Zelig 3.5 does not solve the problem.
> Thanks,
> Antonio
>
>
>
>
>
> -
> --
> Zelig Mailing List, served by HUIT
> Send messages: zelig(a)lists.gking.harvard.edu
> [un]subscribe Options: http://lists.gking.harvard.edu/mailman/listinfo/zelig
> Zelig program information: http://gking.harvard.edu/zelig/
> Zelig mailing list
> Zelig(a)lists.gking.harvard.edu
>
> To unsubscribe from this list or get other information:
>
> https://lists.gking.harvard.edu/mailman/listinfo/zelig
Hi there,
I need to know who matched with who using Genetic matching. I
can get the match.matrix when I use nearest, optimal, or other
methods. But when I set the method as "genetic", the match.matrix I got was
null. Do you know how to get the matched pairs?
Thank you in advance.
Best,
Xin